

*******************************************************************************
*** Description: 	This document provides the code for reproducing the 	***
***					figures in the paper, "Less Human Than Human: Threat,  	***
***					Language, and Relative Dehumanization" which is  		***
***					authored by Shane P. Singh and Jaroslav Tir and appears ***
***					in the British Journal of Political Science.			***
***																			***	
***					It also provides the code for reproducing the tables 	***
***					and figures in the Supplementary Material.				***
***					 														***
***					It also provides the code for reproducing statistics 	***
***					associated with claims made in the text.				***
*******************************************************************************


**************
**************
*Set the Version                                                                                                                                 
**************
**************
version 17.0


**************
**************
*Install Required Package (remove initial asterisk if package not yet installed)                                                                                                                     
**************
**************
*ssc install coefplot




**************
**************
*Change the font for the figures
**************
**************	
graph set window fontface "LMRoman10-Regular"




**************
**************
*Open the Analysis Dataset
**************
**************
use "Singh_Tir_Dehumanization_Replication_Analysis.dta", clear





**************
**************
*Evidence for Claim Made in the Text: "The breakdown per experimental group is: Pulwama Condition, English (275); 
*Pulwama Condition, Hindi (265); China Condition, English (268); China Condition, Hindi (264); 
*Shipping Condition, English (273); Shipping Condition, Hindi (269)."
**************
**************
tab treatment_condition hindi_survey




**************
**************
*Evidence for Claim Made in the Text: "There is no evidence that Muslims somehow feel 'less Indian' than Hindus. 
*Mean responses to a pre-treatment nationalism question that asks about the importance of being Indian do not 
*significantly differ across Hindus and Muslims (two-sided p = .765). 
**************
**************
reg  important_indian_pre i.religion




**************
**************
*Figure 2
**************
**************
preserve 

gen where = _n-11 in 1/121

kdensity evol_Hindus_pre if hindu == 1, bw(3) at(where) gen(x_Hindus d_Hindus)

kdensity evol_Muslims_pre if hindu == 1, bw(3) at(where) gen(x_Muslims d_Muslims)

twoway  ///
	( ///
	area d_Hindus where, ///
	color(orange%65) lwidth(none) ///
	) ///
	( ///
	area d_Muslims where, ///
	color(green%65) lwidth(none) ///
	) ///
		,	xtitle("") ytitle(Probability Density, size(large)) ///
			legend(order(1 "Hindus" 2 "Muslims") pos(6) rows(1) size(large)) ///
			xlabel(0(25)100) xtitle("Humanness Rating", size(large)) ///
			ylabel(0(.01).05) ///
			xscale(titlegap(2)) ///
			xsize(10) ysize(10) ///
			scheme(s1color) ///
			name(density_Hin_Mus, replace)
			
restore 



preserve 

gen where = _n-11 in 1/121

kdensity evol_Ind_pre , bw(3) at(where) gen(x_Indians d_Indians)

kdensity evol_Chinese_pre , bw(3) at(where) gen(x_Chinese d_Chinese)


twoway  ///
	( ///
	area d_Indians where, ///
	color(orange%85) lwidth(none) ///
	) ///
	( ///
	area d_Chinese where, ///
	color(red%65) lwidth(none) ///
	) ///
		,	xtitle("") ytitle(Probability Density, size(large)) ///
			legend(order(1 "Indians" 2 "Chinese") pos(6) rows(1) size(large)) ///
			xlabel(0(25)100) xtitle("Humanness Rating", size(large)) ///
			ylabel(0(.01).05) xscale(titlegap(2)) ///
			xscale(titlegap(2)) ///
			xsize(10) ysize(10) ///
			scheme(s1color)  ///
			name(density_Ind_China, replace)

restore 




*Combine Graphs
graph combine  density_Hin_Mus density_Ind_China ///
	, 	graphregion(margin(zero) color(white))  rows(1)  xsize(7) scale(1) ycommon



**************
**************
*Figure 3
**************
**************	
reg evol_Hindus_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Hindus 

reg evol_Muslims_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Muslims

reg evol_Hindus_rel_to_Mus_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Hindus_rel_to_Mus


reg evol_Hindus_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Hindus_covs

reg evol_Muslims_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Muslims_covs

reg evol_Hindus_rel_to_Mus_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Hindus_rel_to_Mus_covs 



coefplot 	///
			(evol_Hindus,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hindus_covs,  offset(.29) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Muslims,  offset(.06) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Muslims_covs,  offset(-.06) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hindus_rel_to_Mus,  offset(-.29) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hindus_rel_to_Mus_covs,  offset(-.41) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .61 	"Hindus"  ///
				 .95 	"Muslims"  ///
				1.3 	"Hindus Rel. to Muslims"  ///
			, labsize(medsmall) noticks)  ///
		yline(.825 1.175) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "w/o covariates" 4 "w/ covariates"))  /// 
		ysize(1.4) scale(2)	///		
		xtitle("ATE of Being Surveyed in Hindi Relative to English",  size(medium))  ytitle("") 
	

	
	

**************
**************
*Figure 4
**************
**************	
reg evol_Ind_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind 

reg evol_Chinese_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Chinese

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_rel_to_Chin


reg evol_Ind_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_covs

reg evol_Chinese_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Chinese_covs

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_rel_to_Chin_covs 

***
*Create Graph
***
coefplot 	///
			(evol_Ind,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs,  offset(.29) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese,  offset(.06) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs,  offset(-.06) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_Ind_rel_to_Chin,  offset(-.29) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_rel_to_Chin_covs,  offset(-.41) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .61 	"Indians"  ///
				 .95 	"Chinese"  ///
				1.3 	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks)  ///
		yline(.825 1.175) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "w/o covariates" 4 "w/ covariates"))  /// 
		ysize(1.4) scale(2)	///
		xtitle("ATE of Being Surveyed in Hindi Relative to English",  size(medium))  ytitle("") 

		
		
		
**************
**************
*Figure 5
**************
**************	
reg biggest_problem_terrorism  b3.treatment_condition  if  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store SMC_Pulwama

reg biggest_problem_China  b3.treatment_condition  if  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store SMC_China 


coefplot 	///
			(SMC_Pulwama,  offset(0) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
			(SMC_China,  offset(0) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90)	///
		xline(0, lcolor(black) lpattern(dash))		///
		ylabel(	 1.02 		`" "Pr(Terror Threat" "Biggest Problem)"  "'   ///
				2.02		`" "Pr(China Threat" "Biggest Problem)"  "'  ///
			, labsize(medsmall) noticks) grid(none) ///
		xlabel(0(.05).15) xscale(range(-.001 .15) titlegap(2)) ///
		yline(1.5, lwidth(medium) lcolor(black)) ///
		legend(order(2 "Pulwama Condition" 4 "China Condition"))  /// 
		ysize(1.3) xsize(5.24) scale(2.15)	///
		xtitle("ITT Effect of Assignment to Condition",  size(medium))  ytitle("")

		
		
		
	
**************
**************
*Figure 6
**************
**************	

***********
*ITT Effects
***********
reg evol_Hindus_post b3.treatment_condition evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Hin_ITT

reg evol_Muslims_post b3.treatment_condition evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Mus_ITT

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Hi_rel_to_Mu_ITT


reg evol_Hindus_post b3.treatment_condition evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Hin_covs_ITT

reg evol_Muslims_post b3.treatment_condition evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Mus_covs_ITT

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Hi_rel_to_Mu_covs_ITT 




***********
*CACEs
***********
ivregress 2sls evol_Hindus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Hin_CACE 

ivregress 2sls evol_Muslims_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Mus_CACE

ivregress 2sls evol_Hindus_rel_to_Mus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Hi_rel_to_Mu_CACE


ivregress 2sls evol_Hindus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Hin_covs_CACE

ivregress 2sls evol_Muslims_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Mus_covs_CACE

ivregress 2sls evol_Hindus_rel_to_Mus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store evol_Hi_rel_to_Mu_covs_CACE 


***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Hin_ITT,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_ITT,  offset(.27) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_ITT,  offset(-.257) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_ITT,  offset(-.397) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hi_rel_to_Mu_ITT,  offset(-.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hi_rel_to_Mu_covs_ITT,  offset(-1.063) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
			(evol_Hin_CACE,  offset(1.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_CACE,  offset(.923) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_CACE,  offset(0.397) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_CACE,  offset(0.257) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hi_rel_to_Mu_CACE,  offset(-0.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hi_rel_to_Mu_covs_CACE,  offset(-0.41) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .8 	"Hindus"  ///
				 1.465 	"Muslims"  ///
				2.13	"Hindus Rel. to Muslims"  ///
			, labsize(medsmall) noticks)  ///
		yline(1.1667 1.8334) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(1.9) scale(1.47)	 ///		
		xtitle("Effect of Being Assigned to Pulwama Condition",  size(medium))  ytitle("")


		
	
		
**************
**************
*Figure 7
**************
**************	

***********
*ITT Effects
***********
reg evol_Ind_post b3.treatment_condition evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_ITT 

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Chinese_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ch_ITT


reg evol_Ind_post b3.treatment_condition evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_covs_ITT

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Chinese_covs_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ch_covs_ITT


***********
*CACEs
***********
ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Chinese_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ch_CACE


ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_covs_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Chinese_covs_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ch_covs_CACE

***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Ind_ITT,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT,  offset(.27) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT,  offset(-.257) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT,  offset(-.397) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ch_ITT,  offset(-.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ch_covs_ITT,  offset(-1.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE,  offset(1.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE,  offset(.923) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE,  offset(0.397) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE,  offset(0.257) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ch_CACE,  offset(-0.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ch_covs_CACE,  offset(-0.41) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .8 	"Indians"  ///
				 1.465 	"Chinese"  ///
				2.13	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks)  ///
		yline(1.1667 1.8334) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(1.9) scale(1.47)	 ///		
		xtitle("Effect of Being Assigned to China Condition",  size(medium))  ytitle("") 

		
		
**************
**************
*Evidence for Claim Made in the Text: "Compared to the Hindu reference group, on average, the humanness of Muslims is 22.91 points lower, 
*while the humanness of the Chinese is just 15.79 points lower than that of Indians. The difference in these differences is statistically 
*significant (two-sided p < .001)."
**************
**************		
sum evol_Hindus_rel_to_Mus_pre if  hindu == 1
sum evol_Ind_rel_to_Chin_pre 
ttesti    1569    15.78521    31.09111   879    22.90899     31.9691



**************
**************
*Evidence for Claim Made in the Text: "Concerning baseline levels of absolute humanization, on average, the humanness of Muslims is 60.47, 
*while the humanness of the Chinese is 66.47. The difference in these scores is again statistically significant (two-sided p < .001)."
**************
**************		
sum evol_Muslims_pre if hindu == 1
sum evol_Chinese_pre
ttesti   1571     66.4704    29.27695  880    60.46818    29.04355 



		
**************
**************
*Figure 8
**************
**************	

***********
*ITT Effects, By Language
*********** 
reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_ITT_E 
reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_ITT_H 

reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Mus_ITT_E 
reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Mus_ITT_H 

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_rel_ITT_E
reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_rel_ITT_H

reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_covs_ITT_E
reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_covs_ITT_H

reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Mus_covs_ITT_E
reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Mus_covs_ITT_H

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_rel_covs_ITT_E 
reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_rel_covs_ITT_H 


***********
*CACEs, By Language
*********** 
reg FMC_TR_passed_pulwama evol_Hindus_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_CACE_E
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_CACE_H
drop yhat

reg FMC_TR_passed_pulwama evol_Muslims_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Mus_CACE_E 
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Mus_CACE_H 
drop yhat

reg FMC_TR_passed_pulwama evol_Hindus_rel_to_Mus_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_rel_CACE_E
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_rel_CACE_H
drop yhat


reg FMC_TR_passed_pulwama evol_Hindus_pre  b3.treatment_condition hindi_survey woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_covs_CACE_E 
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_covs_CACE_H 
drop yhat

reg FMC_TR_passed_pulwama evol_Muslims_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Mus_covs_CACE_E
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Mus_covs_CACE_H
drop yhat

reg FMC_TR_passed_pulwama evol_Hindus_rel_to_Mus_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Hin_rel_covs_CACE_E
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. &  hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Hin_rel_covs_CACE_H
drop yhat


***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Hin_ITT_E,  offset(1.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_ITT_E,  offset(1.27) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_ITT_E,  offset(0.743) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_ITT_E,  offset(0.603) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hin_rel_ITT_E,  offset(0.077) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_rel_covs_ITT_E,  offset(-0.063) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
			(evol_Hin_CACE_E,  offset(2.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_CACE_E,  offset(1.923) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_CACE_E,  offset(1.396) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_CACE_E,  offset(1.256) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hin_rel_CACE_E,  offset(0.730) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_rel_covs_CACE_E,  offset(0.590) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///
			(evol_Hin_ITT_H,  offset(-0.590) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_ITT_H,  offset(-0.730) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_ITT_H,  offset(-1.256) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_ITT_H,  offset(-1.396) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hin_rel_ITT_H,  offset(-1.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_rel_covs_ITT_H,  offset(-2.063) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
			(evol_Hin_CACE_H,  offset(0.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_covs_CACE_H,  offset(-0.077) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(evol_Mus_CACE_H,  offset(-0.603) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Mus_covs_CACE_H,  offset(-0.743) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///		
			(evol_Hin_rel_CACE_H,  offset(-1.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Hin_rel_covs_CACE_H,  offset(-1.41) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		yaxis(1 2)  ytitle("", axis(1)) ytitle("", axis(2))	///
		yscale(range(-.5 3.5)) ///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 -.21 	"Hindus"  ///
				 .457 	"Muslims"  ///
				1.123 	"Hindus Rel. to Muslims"  ///
				1.790 	"Hindus"  ///
				2.457 	"Muslims"  ///
				3.123 	"Hindus Rel. to Muslims"  ///
			, labsize(medsmall) noticks axis(2) angle(0))  ///
		ylabel(	 .5	`" "{bf:English}"  "'  ///
				 2.5  `" "{bf:Hindi}" "' ///
			, labsize(medsmall) noticks axis(1) angle(270))  ///
		yline(.167 .833 2.167 2.833) ///
		yline(1.5, lwidth(medthick) lcolor(black)) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(2.75) scale(.97)	///		
		xtitle("Effect of Being Assigned to Pulwama Condition",  size(medium))  ytitle("") 	
	
	
	
	
**************
**************
*Figure 9
**************
**************	

***********
*ITT Effects, By Language
*********** 
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_ITT_E 
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_ITT_H 

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_ITT_E 
reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_ITT_H 

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_ITT_E
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_ITT_H

reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_covs_ITT_E
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_covs_ITT_H

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_covs_ITT_E
reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_covs_ITT_H

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_covs_ITT_E 
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_covs_ITT_H 



***********
*CACEs, By Language
*********** 
reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_CACE_E
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_CACE_H
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_CACE_E 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_CACE_H 
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_CACE_E
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_CACE_H
drop yhat


reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_covs_CACE_E 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_covs_CACE_H 
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_covs_CACE_E
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_covs_CACE_H
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_covs_CACE_E
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_covs_CACE_H
drop yhat



***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Ind_ITT_E,  offset(1.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT_E,  offset(1.27) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT_E,  offset(0.743) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT_E,  offset(0.603) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_ITT_E,  offset(0.077) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_ITT_E,  offset(-0.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE_E,  offset(2.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE_E,  offset(1.923) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE_E,  offset(1.396) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE_E,  offset(1.256) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_CACE_E,  offset(0.730) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_CACE_E,  offset(0.590) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///
			(evol_Ind_ITT_H,  offset(-0.590) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT_H,  offset(-0.730) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT_H,  offset(-1.256) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT_H,  offset(-1.396) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_ITT_H,  offset(-1.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_ITT_H,  offset(-2.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE_H,  offset(0.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE_H,  offset(-0.077) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE_H,  offset(-0.603) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE_H,  offset(-0.743) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_CACE_H,  offset(-1.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_CACE_H,  offset(-1.41) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		yaxis(1 2)  ytitle("", axis(1)) ytitle("", axis(2))	///
		yscale(range(-.5 3.5)) ///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 -.21 	"Indians"  ///
				 .457 	"Chinese"  ///
				1.123 	"Indians Rel. to Chinese"  ///
				1.790 	"Indians"  ///
				2.457 	"Chinese"  ///
				3.123 	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks axis(2) angle(0))  ///
		ylabel(	 .5	`" "{bf:English}"  "'  ///
				 2.5  `" "{bf:Hindi}" "' ///
			, labsize(medsmall) noticks axis(1) angle(270))  ///
		yline(.167 .833 2.167 2.833) ///
		yline(1.5, lwidth(medthick) lcolor(black)) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(2.75) scale(.97)	///		
		xtitle("Effect of Being Assigned to China Condition",  size(medium))  ytitle("")	
	
	
	
	
	
**************
**************
*Evidence for Claim Made in the Text: "	For those taking the survey in English, the effect of assignment to the China threat condition on relative dehumanization is 1.53 *(two-sided p = .378) overall and 4.41 (two-sided p = .303) for those who were effectively manipulated by the treatment. For those in the Hindi setting, the effect is *3.05 (two-sided p = .080) overall and 7.11 (two-sided p = .086) for those who were effectively manipulated"
**************
**************		
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0))
	
reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0))	
drop yhat
	
	
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1))
	
reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=. &   treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1))	
drop yhat
	
	
	
**************
**************
*Table SM1
**************
**************
reg evol_Hindus_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Hindus 

reg evol_Muslims_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Muslims

reg evol_Hindus_rel_to_Mus_pre i.hindi_survey  if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Hindus_rel_to_Mus


reg evol_Hindus_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Hindus_covs

reg evol_Muslims_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Muslims_covs

reg evol_Hindus_rel_to_Mus_pre i.hindi_survey  woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=.  & hindu == 1 
estimates store evol_Hindus_rel_to_Mus_covs 

preserve
label var evol_Hindus_pre "Hindus"
label var evol_Muslims_pre "Muslims"
label var evol_Hindus_rel_to_Mus_pre "Hindus rel. Muslims"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Hindus ///
evol_Muslims ///
evol_Hindus_rel_to_Mus ///
evol_Hindus_covs ///
evol_Muslims_covs ///
evol_Hindus_rel_to_Mus_covs ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Language on Muslim Dehumanization)
restore


**************
**************
*Table SM2
**************
**************	
reg evol_Ind_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Ind 

reg evol_Chinese_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Chinese

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Ind_rel_to_Chin


reg evol_Ind_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Ind_covs

reg evol_Chinese_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Chinese_covs

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  
estimates store evol_Ind_rel_to_Chin_covs 

preserve
label var evol_Ind_pre "Indians"
label var evol_Chinese_pre "Chinese"
label var evol_Ind_rel_to_Chin_pre "Indians rel. Chinese"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Ind ///
evol_Chinese ///
evol_Ind_rel_to_Chin ///
evol_Ind_covs ///
evol_Chinese_covs ///
evol_Ind_rel_to_Chin_covs ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Language on Chinese Dehumanization)
restore

	
		
**************
**************
*Table SM3
**************
**************
reg biggest_problem_China  b3.treatment_condition   if  treatment_condition~=1
estimates store SMC_China 

reg biggest_problem_terrorism  b3.treatment_condition   if  treatment_condition~=2
estimates store SMC_Pulwama


preserve
label var biggest_problem_China "China Threat Biggest Problem"
label var biggest_problem_terrorism "Terror Threat Biggest Problem"
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

esttab ///
SMC_China ///
SMC_Pulwama ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Vignette Assignment on Perceptions of Threat)
restore



**************
**************
*Table SM4
**************
**************	
reg evol_Hindus_post b3.treatment_condition evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_ITT

reg evol_Muslims_post b3.treatment_condition evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_ITT

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hi_rel_to_Mu_ITT


reg evol_Hindus_post b3.treatment_condition evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_covs_ITT

reg evol_Muslims_post b3.treatment_condition evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_covs_ITT

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hi_rel_to_Mu_covs_ITT 



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var evol_Hindus_post "Hindus"
label var evol_Muslims_post "Muslims"
label var evol_Hindus_rel_to_Mus_post "Hindus rel. Muslims"
label var evol_Hindus_pre "Hindus (pre-treatment)"
label var evol_Muslims_pre "Muslims (pre-treatment)"
label var evol_Hindus_rel_to_Mus_pre "Hindus rel. Muslims (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Hin_ITT ///
evol_Mus_ITT ///
evol_Hi_rel_to_Mu_ITT ///
evol_Hin_covs_ITT ///
evol_Mus_covs_ITT ///
evol_Hi_rel_to_Mu_covs_ITT ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Muslim Dehumanization (ITT Effects))
restore	
		
		
		
**************
**************
*Table SM5
**************
**************
ivregress 2sls evol_Hindus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_CACE 

ivregress 2sls evol_Muslims_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_CACE

ivregress 2sls evol_Hindus_rel_to_Mus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hi_rel_to_Mu_CACE


ivregress 2sls evol_Hindus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_covs_CACE

ivregress 2sls evol_Muslims_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_covs_CACE

ivregress 2sls evol_Hindus_rel_to_Mus_post (FMC_TR_passed_pulwama = b3.treatment_condition) evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=. & hindu == 1 &  treatment_condition~=2
estimates store evol_Hi_rel_to_Mu_covs_CACE 



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var FMC_TR_passed_pulwama "Passed FMC  (instrumented)"
label var evol_Hindus_post "Hindus"
label var evol_Muslims_post "Muslims"
label var evol_Hindus_rel_to_Mus_post "Hindus rel. Muslims"
label var evol_Hindus_pre "Hindus (pre-treatment)"
label var evol_Muslims_pre "Muslims (pre-treatment)"
label var evol_Hindus_rel_to_Mus_pre "Hindus rel. Muslims (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Hin_CACE ///
evol_Mus_CACE ///
evol_Hi_rel_to_Mu_CACE ///
evol_Hin_covs_CACE ///
evol_Mus_covs_CACE ///
evol_Hi_rel_to_Mu_covs_CACE ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Muslim Dehumanization (CACEs))
restore



	


**************
**************
*Table SM6
**************
**************	
reg evol_Ind_post b3.treatment_condition evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_ITT 

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_to_Ch_ITT


reg evol_Ind_post b3.treatment_condition evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_covs_ITT

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_covs_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_to_Ch_covs_ITT



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var evol_Ind_post "Indians"
label var evol_Chinese_post "Chinese"
label var evol_Ind_rel_to_Chin_post "Indians rel. Chinese"
label var evol_Ind_pre "Indians (pre-treatment)"
label var evol_Chinese_pre "Chinese (pre-treatment)"
label var evol_Ind_rel_to_Chin_pre "Indians rel. Chinese (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Ind_ITT ///
evol_Chinese_ITT ///
evol_In_rel_to_Ch_ITT ///
evol_Ind_covs_ITT ///
evol_Chinese_covs_ITT ///
evol_In_rel_to_Ch_covs_ITT ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Chinese Dehumanization (ITT Effects))
restore	
		
**************
**************
*Table SM7
**************
**************
ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_to_Ch_CACE


ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_covs_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_covs_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_to_Ch_covs_CACE



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var FMC_TR_passed_China "Passed FMC (instrumented)"
label var evol_Ind_post "Indians"
label var evol_Chinese_post "Chinese"
label var evol_Ind_rel_to_Chin_post "Indians rel. Chinese"
label var evol_Ind_pre "Indians (pre-treatment)"
label var evol_Chinese_pre "Chinese (pre-treatment)"
label var evol_Ind_rel_to_Chin_pre "Indians rel. Chinese (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Ind_CACE ///
evol_Chinese_CACE ///
evol_In_rel_to_Ch_CACE ///
evol_Ind_covs_CACE ///
evol_Chinese_covs_CACE ///
evol_In_rel_to_Ch_covs_CACE ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Chinese Dehumanization (CACEs))
restore




**************
**************
*Table SM8
**************
**************	
reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_ITT 


reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_ITT 


reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_rel_ITT


reg evol_Hindus_post b3.treatment_condition##i.hindi_survey evol_Hindus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_covs_ITT


reg evol_Muslims_post b3.treatment_condition##i.hindi_survey evol_Muslims_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Mus_covs_ITT

reg evol_Hindus_rel_to_Mus_post b3.treatment_condition##i.hindi_survey evol_Hindus_rel_to_Mus_pre woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2
estimates store evol_Hin_rel_covs_ITT 




preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var evol_Hindus_post "Hindus"
label var evol_Muslims_post "Muslims"
label var evol_Hindus_rel_to_Mus_post "Hindus rel. Muslims"
label var evol_Hindus_pre "Hindus (pre-treatment)"
label var evol_Muslims_pre "Muslims (pre-treatment)"
label var evol_Hindus_rel_to_Mus_pre "Hindus rel. Muslims (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Hin_ITT ///
evol_Mus_ITT ///
evol_Hin_rel_ITT ///
evol_Hin_covs_ITT ///
evol_Mus_covs_ITT ///
evol_Hin_rel_covs_ITT ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Muslim Dehumanization by Language (ITT Effects))
restore


		
**************
**************
*Table SM9
**************
**************
reg FMC_TR_passed_pulwama evol_Hindus_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Hin_CACE
drop yhat

reg FMC_TR_passed_pulwama evol_Muslims_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Mus_CACE 
drop yhat


reg FMC_TR_passed_pulwama evol_Hindus_rel_to_Mus_pre  b3.treatment_condition hindi_survey  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre    if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Hin_rel_CACE
drop yhat



reg FMC_TR_passed_pulwama evol_Hindus_pre  b3.treatment_condition hindi_survey woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_post c.yhat##i.hindi_survey evol_Hindus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Hin_covs_CACE 
drop yhat

reg FMC_TR_passed_pulwama evol_Muslims_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Muslims_post c.yhat##i.hindi_survey evol_Muslims_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Mus_covs_CACE
drop yhat

reg FMC_TR_passed_pulwama evol_Hindus_rel_to_Mus_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2 
predict yhat,  xb 
reg evol_Hindus_rel_to_Mus_post c.yhat##i.hindi_survey evol_Hindus_rel_to_Mus_pre  woman age_10 income education  if evol_Hindus_rel_to_Mus_pre~=. & evol_Hindus_rel_to_Mus_post~=.  & hindu == 1 &  treatment_condition~=2, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Hin_rel_covs_CACE
drop yhat



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

gen yhat = .
label var yhat "Passed FMC (instrumented)"
label var evol_Hindus_post "Hindus"
label var evol_Muslims_post "Muslims"
label var evol_Hindus_rel_to_Mus_post "Hindus rel. Muslims"
label var evol_Hindus_pre "Hindus (pre-treatment)"
label var evol_Muslims_pre "Muslims (pre-treatment)"
label var evol_Hindus_rel_to_Mus_pre "Hindus rel. Muslims (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Hin_CACE ///
evol_Mus_CACE ///
evol_Hin_rel_CACE ///
evol_Hin_covs_CACE ///
evol_Mus_covs_CACE ///
evol_Hin_rel_covs_CACE ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Muslim Dehumanization by Language (CACEs))
restore




**************
**************
*Table SM10
**************
**************	

reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_ITT

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_ITT


reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Ind_covs_ITT

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_Chinese_covs_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1
estimates store evol_In_rel_covs_ITT



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

label var evol_Ind_post "Indians"
label var evol_Chinese_post "Chinese"
label var evol_Ind_rel_to_Chin_post "Indians rel. Chinese"
label var evol_Ind_pre "Indians (pre-treatment)"
label var evol_Chinese_pre "Chinese (pre-treatment)"
label var evol_Ind_rel_to_Chin_pre "Indians rel. Chinese (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Ind_ITT ///
evol_Chinese_ITT ///
evol_In_rel_ITT ///
evol_Ind_covs_ITT ///
evol_Chinese_covs_ITT ///
evol_In_rel_covs_ITT ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Chinese Dehumanization by Language (ITT Effects))
restore


	
	


**************
**************
*Table SM11
**************
**************	
reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Ind_CACE
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Chinese_CACE 
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_In_rel_CACE
drop yhat


reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Ind_covs_CACE
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_Chinese_covs_CACE
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
estimates store evol_In_rel_covs_CACE
drop yhat



preserve
label define treatment 1 "Pulwama Condition", modify
label define treatment 2 "China Condition", modify

gen yhat = .
label var yhat "Passed FMC (instrumented)"
label var evol_Ind_post "Indians"
label var evol_Chinese_post "Chinese"
label var evol_Ind_rel_to_Chin_post "Indians rel. Chinese"
label var evol_Ind_pre "Indians (pre-treatment)"
label var evol_Chinese_pre "Chinese (pre-treatment)"
label var evol_Ind_rel_to_Chin_pre "Indians rel. Chinese (pre-treatment)"
label var hindi_survey "Hindi"
label var woman "Woman"
label var age_10 "Age"
label var income "Income"
label var education "Education"

esttab ///
evol_Ind_CACE ///
evol_Chinese_CACE ///
evol_In_rel_CACE ///
evol_Ind_covs_CACE ///
evol_Chinese_covs_CACE ///
evol_In_rel_covs_CACE ///
,  b(3) se(3)  nostar tex r2 nobaselevels label title(Estimated Effects of Threat on Chinese Dehumanization by Language (CACEs))
restore


	

**************
**************
*Figure SM1
**************
**************
reg evol_Ind_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind 

reg evol_Chinese_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Chinese

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_rel_to_Chin


reg evol_Ind_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_covs

reg evol_Chinese_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Chinese_covs

reg evol_Ind_rel_to_Chin_pre i.hindi_survey  woman age_10 income education if evol_Ind_rel_to_Chin_pre~=.  & hindu == 1 
margins, dydx(hindi_survey) post coeflegend
estimates store evol_Ind_rel_to_Chin_covs 

***
*Create Graph
***


coefplot 	///
			(evol_Ind,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs,  offset(.29) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese,  offset(.06) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs,  offset(-.06) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_Ind_rel_to_Chin,  offset(-.29) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_rel_to_Chin_covs,  offset(-.41) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .61 	"Indians"  ///
				 .95 	"Chinese"  ///
				1.3 	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks)  ///
		yline(.825 1.175) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "w/o covariates" 4 "w/ covariates"))  /// 
		ysize(1.4) scale(2)	///
		xtitle("ATE of Being Surveyed in Hindi Relative to English",  size(medium))  ytitle("")






**************
**************
*Figure SM2
**************
**************

***********
*ITT Effects
***********

reg evol_Ind_post b3.treatment_condition evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_ITT 

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Chinese_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ch_ITT


reg evol_Ind_post b3.treatment_condition evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_covs_ITT

reg evol_Chinese_post b3.treatment_condition evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Chinese_covs_ITT

reg evol_Ind_rel_to_Chin_post b3.treatment_condition evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ch_covs_ITT


***********
*CACEs
***********
ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Chinese_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ch_CACE


ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_covs_CACE

ivregress 2sls evol_Chinese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Chinese_covs_CACE

ivregress 2sls evol_Ind_rel_to_Chin_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ch_covs_CACE

***
*Create Graph of ITT Effects and CACEs
***

coefplot 	///
			(evol_Ind_ITT,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT,  offset(.27) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT,  offset(-.257) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT,  offset(-.397) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ch_ITT,  offset(-.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ch_covs_ITT,  offset(-1.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE,  offset(1.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE,  offset(.923) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE,  offset(0.397) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE,  offset(0.257) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ch_CACE,  offset(-0.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ch_covs_CACE,  offset(-0.41) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .8 	"Indians"  ///
				 1.465 	"Chinese"  ///
				2.13	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks)  ///
		yline(1.1667 1.8334) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(1.9) scale(1.47)	 ///		
		xtitle("Effect of Being Assigned to China Condition",  size(medium))  ytitle("")		




**************
**************
*Figure SM3
**************
**************

***********
*ITT Effects, By Language
*********** 
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_ITT_E 
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_ITT_H 

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_ITT_E 
reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_ITT_H 

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_ITT_E
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_ITT_H

reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_covs_ITT_E
reg evol_Ind_post b3.treatment_condition##i.hindi_survey evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_covs_ITT_H

reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_covs_ITT_E
reg evol_Chinese_post b3.treatment_condition##i.hindi_survey evol_Chinese_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_covs_ITT_H

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_covs_ITT_E 
reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_covs_ITT_H 

reg evol_Ind_rel_to_Chin_post b3.treatment_condition##i.hindi_survey evol_Ind_rel_to_Chin_pre woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1
margins, dydx(treatment_condition) at(hindi_survey = (0 1))

***********
*CACEs, By Language
*********** 
reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_CACE_E
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_CACE_H
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_CACE_E 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_CACE_H 
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_CACE_E
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre    if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_CACE_H
drop yhat


reg FMC_TR_passed_China evol_Ind_pre  b3.treatment_condition hindi_survey woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Ind_covs_CACE_E 
reg evol_Ind_post c.yhat##i.hindi_survey evol_Ind_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Ind_covs_CACE_H 
drop yhat

reg FMC_TR_passed_China evol_Chinese_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_Chinese_covs_CACE_E
reg evol_Chinese_post c.yhat##i.hindi_survey evol_Chinese_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_Chinese_covs_CACE_H
drop yhat

reg FMC_TR_passed_China evol_Ind_rel_to_Chin_pre  b3.treatment_condition hindi_survey woman age_10 income education if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1 
predict yhat,  xb 
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0)) post coeflegend
estimates store evol_In_rel_covs_CACE_E
reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (1)) post coeflegend
estimates store evol_In_rel_covs_CACE_H

reg evol_Ind_rel_to_Chin_post c.yhat##i.hindi_survey evol_Ind_rel_to_Chin_pre  woman age_10 income education  if evol_Ind_rel_to_Chin_pre~=. & evol_Ind_rel_to_Chin_post~=.  & hindu == 1 &  treatment_condition~=1, vce(bootstrap, reps(250) seed(1234)) 
margins, dydx(yhat) at(hindi_survey = (0 1))


drop yhat



***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Ind_ITT_E,  offset(1.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT_E,  offset(1.27) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT_E,  offset(0.743) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT_E,  offset(0.603) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_ITT_E,  offset(0.077) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_ITT_E,  offset(-0.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE_E,  offset(2.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE_E,  offset(1.923) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE_E,  offset(1.396) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE_E,  offset(1.256) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_CACE_E,  offset(0.730) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_CACE_E,  offset(0.590) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///
			(evol_Ind_ITT_H,  offset(-0.590) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT_H,  offset(-0.730) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_ITT_H,  offset(-1.256) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_ITT_H,  offset(-1.396) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_ITT_H,  offset(-1.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_ITT_H,  offset(-2.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE_H,  offset(0.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE_H,  offset(-0.077) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Chinese_CACE_H,  offset(-0.603) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Chinese_covs_CACE_H,  offset(-0.743) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_CACE_H,  offset(-1.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_covs_CACE_H,  offset(-1.41) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		yaxis(1 2)  ytitle("", axis(1)) ytitle("", axis(2))	///
		yscale(range(-.5 3.5)) ///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 -.21 	"Indians"  ///
				 .457 	"Chinese"  ///
				1.123 	"Indians Rel. to Chinese"  ///
				1.790 	"Indians"  ///
				2.457 	"Chinese"  ///
				3.123 	"Indians Rel. to Chinese"  ///
			, labsize(medsmall) noticks axis(2) angle(0))  ///
		ylabel(	 .5	`" "{bf:English}"  "'  ///
				 2.5  `" "{bf:Hindi}" "' ///
			, labsize(medsmall) noticks axis(1) angle(270))  ///
		yline(.167 .833 2.167 2.833) ///
		yline(1.5, lwidth(medthick) lcolor(black)) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(2.75) scale(.97)	///		
		xtitle("Effect of Being Assigned to China Condition",  size(medium))  ytitle("")
		
		
		


**************
**************
*Figure SM4
**************
**************

***********
*ITT Effects
***********
reg important_indian_post b3.treatment_condition important_indian_pre if  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store impor_ind_ITT 

reg important_indian_post b3.treatment_condition important_indian_pre woman age_10 income education if  treatment_condition~=2
margins, dydx(treatment_condition) post coeflegend
estimates store impor_ind_covs_ITT



***********
*CACEs
***********
ivregress 2sls important_indian_post (FMC_TR_passed_pulwama = b3.treatment_condition) important_indian_pre if  treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store impor_ind_CACE

ivregress 2sls important_indian_post (FMC_TR_passed_pulwama = b3.treatment_condition) important_indian_pre woman age_10 income education if treatment_condition~=2
margins, dydx(FMC_TR_passed_pulwama) post coeflegend
estimates store impor_ind_covs_CACE



***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(impor_ind_ITT,  offset(.15) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(impor_ind_covs_ITT,  offset(-.15) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(impor_ind_CACE,  offset(.15) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(impor_ind_covs_CACE,  offset(-.15) msymbol(square) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	"" ///
			, labsize(medsmall) noticks)  ///
		xlabel(-.75(.25).75)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 6 "CACE, w/o covariates" 4 "ITT, w/ covariates"  8 "CACE, w/ covariates") rows(2))  /// 
		ysize(1) xsize(3.5) scale(2)	 ///		
		xtitle("Effect of Being Assigned to Pulwama Condition on Importance of Being Indian",  size(medium))  ytitle("")


		

**************
**************
*Figure SM5
**************
**************

***********
*ITT Effects
***********
reg important_indian_post b3.treatment_condition important_indian_pre if    treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store impor_ind_ITT 

reg important_indian_post b3.treatment_condition important_indian_pre woman age_10 income education if    treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store impor_ind_covs_ITT



***********
*CACEs
***********
ivregress 2sls important_indian_post (FMC_TR_passed_China = b3.treatment_condition) important_indian_pre if    treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store impor_ind_CACE

ivregress 2sls important_indian_post (FMC_TR_passed_China = b3.treatment_condition) important_indian_pre woman age_10 income education if treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store impor_ind_covs_CACE



***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(impor_ind_ITT,  offset(.2) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(impor_ind_covs_ITT,  offset(-.2) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(impor_ind_CACE,  offset(.2) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(impor_ind_covs_CACE,  offset(-.2) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	"" ///
			, labsize(medsmall) noticks)  ///
		xlabel(-.75(.25).75)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 6 "CACE, w/o covariates" 4 "ITT, w/ covariates"  8 "CACE, w/ covariates") rows(2))  /// 
		ysize(1) xsize(3.5) scale(2)	 ///		
		xtitle("Effect of Being Assigned to China Condition on Importance of Being Indian",  size(medium))  ytitle("")




**************
**************
*Figure SM6
**************
**************

***********
*ITT Effects
***********
reg evol_Ind_post b3.treatment_condition evol_Ind_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_ITT 

reg evol_Taiwanese_post b3.treatment_condition evol_Taiwanese_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Taiwanese_ITT

reg evol_Ind_rel_to_Taiw_post b3.treatment_condition evol_Ind_rel_to_Taiw_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ta_ITT


reg evol_Ind_post b3.treatment_condition evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Ind_covs_ITT

reg evol_Taiwanese_post b3.treatment_condition evol_Taiwanese_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_Taiwanese_covs_ITT

reg evol_Ind_rel_to_Taiw_post b3.treatment_condition evol_Ind_rel_to_Taiw_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(treatment_condition) post coeflegend
estimates store evol_In_rel_to_Ta_covs_ITT


***********
*CACEs
***********
ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_CACE

ivregress 2sls evol_Taiwanese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Taiwanese_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Taiwanese_CACE

ivregress 2sls evol_Ind_rel_to_Taiw_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Taiw_pre if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ta_CACE


ivregress 2sls evol_Ind_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Ind_covs_CACE

ivregress 2sls evol_Taiwanese_post (FMC_TR_passed_China = b3.treatment_condition) evol_Taiwanese_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_Taiwanese_covs_CACE

ivregress 2sls evol_Ind_rel_to_Taiw_post (FMC_TR_passed_China = b3.treatment_condition) evol_Ind_rel_to_Taiw_pre woman age_10 income education if evol_Ind_rel_to_Taiw_pre~=. & evol_Ind_rel_to_Taiw_post~=.  &  treatment_condition~=1
margins, dydx(FMC_TR_passed_China) post coeflegend
estimates store evol_In_rel_to_Ta_covs_CACE

***
*Create Graph of ITT Effects and CACEs
***
coefplot 	///
			(evol_Ind_ITT,  offset(.41) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_ITT,  offset(.27) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Taiwanese_ITT,  offset(-.257) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Taiwanese_covs_ITT,  offset(-.397) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ta_ITT,  offset(-.923) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ta_covs_ITT,  offset(-1.063) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
			(evol_Ind_CACE,  offset(1.063) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Ind_covs_CACE,  offset(.923) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(evol_Taiwanese_CACE,  offset(0.397) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_Taiwanese_covs_CACE,  offset(0.257) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///		
			(evol_In_rel_to_Ta_CACE,  offset(-0.27) msymbol(square) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(evol_In_rel_to_Ta_covs_CACE,  offset(-0.41) msymbol(square) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///	
		, scheme(s1color)  title("") levels(90) grid(none)	///
		xline(0, lcolor(black) lpattern(dash))  ///
		ylabel(	 .8 	"Indians"  ///
				 1.465 	"Taiwanese"  ///
				2.13	"Indians Rel. to Taiwanese"  ///
			, labsize(medsmall) noticks)  ///
		yline(1.1667 1.8334) ///
		xlabel(-15(5)15)  ///
		xscale(titlegap(2)) ///
		legend(order(2 "ITT, w/o covariates" 14 "CACE, w/o covariates" 4 "ITT, w/ covariates"  16 "CACE, w/ covariates") rows(2))  /// 
		ysize(1.9) scale(1.47)	 ///		
		xtitle("Effect of Being Assigned to China Condition",  size(medium))  ytitle("") 

		
		
**************
**************
*Figures SM7-SM9
**************
**************


***
*Open the Balance Dataset
***
use "Singh_Tir_Dehumanization_Replication_Balance.dta", clear



***
*Put covariates on a standard deviation scale
***
sum age_10 				
gen age_sd = age_10/r(sd) 

sum income 			
gen income_sd = income/r(sd)

sum education	
gen education_sd = education/r(sd)



***
*Figure SM7
***
reg hindi_survey woman 
margins,  dydx(woman)  post coeflegend
estimates store woman

reg hindi_survey age_sd 
margins,  dydx(age_sd)    post coeflegend
estimates store age_sd

reg  hindi_survey income_sd 
margins,  dydx(income_sd)   post coeflegend
estimates store income_sd

reg hindi_survey education_sd  
margins,  dydx(education_sd)    post coeflegend
estimates store education_sd


coefplot 	///
			(woman,  offset(0) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(age_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(income_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
			(education_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(black) ciopts(color(black*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90)		///
		xline(0, lcolor(black*.8) lpattern(dash))  ylabel( , labsize(medsmall))  xlabel(-.15(.05).15) xscale(range(-.17 .17)) ///
		xscale(titlegap(2)) ///
		legend(off) ysize(1.5) scale(1.7)	///
		xtitle("Estimated Effect of Covariate" "on Pr(Hindi Survey)", size(medium))  ytitle("") 	///
		coeflabels(	woman  = 		"Woman"   ///
					age_sd  = 			"Age"   ///
					income_sd  = 			"Income Level"   ///
					education_sd  = 		"Education Level"   ///
					) 								


***
*Figure SM8
***
reg pulwama woman if treatment_condition ~=2 
margins,  dydx(woman)  post coeflegend
estimates store woman

reg pulwama age_sd if treatment_condition ~=2  
margins,  dydx(age_sd)    post coeflegend
estimates store age_sd

reg  pulwama income_sd if treatment_condition ~=2 
margins,  dydx(income_sd)   post coeflegend
estimates store income_sd

reg pulwama education_sd  if treatment_condition ~=2 
margins,  dydx(education_sd)    post coeflegend
estimates store education_sd


coefplot 	///
			(woman,  offset(0) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(age_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(income_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
			(education_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(green) ciopts(color(green*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90)		///
		xline(0, lcolor(black*.8) lpattern(dash))  ylabel( , labsize(medsmall))  xlabel(-.15(.05).15) xscale(range(-.17 .17)) ///
		xscale(titlegap(2)) ///
		legend(off) ysize(1.5) scale(1.7)	///
		xtitle("Estimated Effect of Covariate" "on Pr(Pulwama Condition)", size(medium))  ytitle("") 	///
		coeflabels(	woman  = 		"Woman"   ///
					age_sd  = 			"Age"   ///
					income_sd  = 			"Income Level"   ///
					education_sd  = 		"Education Level"   ///
					) 	




***
*Figure SM9
***
reg china woman if treatment_condition ~=1 
margins,  dydx(woman)  post coeflegend
estimates store woman

reg china age_sd if treatment_condition ~=1  
margins,  dydx(age_sd)    post coeflegend
estimates store age_sd

reg  china income_sd if treatment_condition ~=1 
margins,  dydx(income_sd)   post coeflegend
estimates store income_sd

reg china education_sd  if treatment_condition ~=1 
margins,  dydx(education_sd)    post coeflegend
estimates store education_sd


coefplot 	///
			(woman,  offset(0) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(age_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(income_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
			(education_sd,  offset(0) msymbol(circle) msize(medlarge)  mcolor(red) ciopts(color(red*.8) lwidth(medium))) ///			
		, scheme(s1color)  title("") levels(90)		///
		xline(0, lcolor(black*.8) lpattern(dash))  ylabel( , labsize(medsmall))  xlabel(-.15(.05).15) xscale(range(-.17 .17)) ///
		xscale(titlegap(2)) ///
		legend(off) ysize(1.5) scale(1.7)	///
		xtitle("Estimated Effect of Covariate" "on Pr(China Condition)", size(medium))  ytitle("") 	///
		coeflabels(	woman  = 		"Woman"   ///
					age_sd  = 			"Age"   ///
					income_sd  = 			"Income Level"   ///
					education_sd  = 		"Education Level"   ///
					) 				
				
				
				
		

**************
**************
*Change the font back to Stata's default
**************
**************	
graph set window fontface "Helvetica" 





























***                     ,__
***                     |  `'.
***  __           |`-._/_.:---`-.._
***  \='.       _/..--'`__         `'-._
***   \- '-.--"`      ===        /   o  `',
***    )= (                 .--_ |       _.'   Thanks for replicating!
***   /_=.'-._             {=_-_ |   .--`-.  
***  /_.'    `\`'-._        '-=   \    _.'
***           )  _.-'`'-..       _..-'`
***          /_.'        `/";';`|
***                       \` .'/
***                        '--'

